After falling 13.7 per cent peak-to-trough during the GFC (following a Steve Keen-style implosion of the banking system), UK house prices have appreciated by a firm 7.5 per cent according to the latest FT/Academetrics index through to January 2010. The Halifax house price index, which is based on the HBOS loan book, has risen by a more sprightly 9.2 per cent in the period to end December 2009 (see first chart below).
In the US, the First American Core Logic index is up 5.2 per cent from its lows, which is near identical to the 5 per cent growth registered by the Case-Shiller index, which while employing a similar repeat-sales methodology is based on a smaller sample of US sales (see next chart). These two indices experienced 32 per cent and 33 per cent peak-to-trough falls, respectively. A third repeat-sales index, produced by the regulator of Fannie Mae and Freddie Mac, which only includes ‘prime’ borrowers, experienced a much smaller 12 per cent peak-to-trough fall. This reinforces the adverse impact of the US sub-prime market on measured house price performance in comparison to, say, the UK, which did not suffer to the same extent from these structural dysfunctions.
Following the release of the December Quarter National Accounts, Rismark estimates that average Australian home prices across all metro and non-metro regions (including all property types) are 4.6 times average Australian disposable household incomes. This is slightly higher than Australia’s average home price-to-income ratio since March 2003 of 4.4 times (see chart). 
For the first time, Rismark has been able to accurately calculate both an ‘average’* and a ‘median’ Australian home price-to-income ratio across all regions and property types. The ratio of median prices to average incomes, which is what we last reported using the September quarter National Accounts data, is a slightly lower 4.3 times (it was 4.1 times in September), as one would expect given skew in the distribution of property values.The new ratios are an important development: previously Rismark and the Reserve Bank of Australia (RBA) had to compare ‘median’ homes prices to ‘average’ disposable household incomes due to the absence of timely quarterly median income data (dividing the ABS National Accounts income data by the number of households only furnishes an average).
In the December quarter, the average Australian home price was $428,298. The median home price was a marginally lower $400,000.
During the GFC, Australia’s average home price-to-income ratio fell to a low of 3.9 as dwelling prices declined while household incomes remained surprisingly stable (care of the government’s fiscal stimulus and a peak unemployment rate of just 5.8 per cent). However, the circa 11 per cent growth in dwelling prices since the start of 2009 has seen the ratio of prices to incomes restored back to around its recent average of 4.4 times.
The cost of Australian housing is a topical subject for the public and the media. In the past, people have made many egregious mistakes when trying to work out the ratio of home prices to incomes. These errors have included:
* using ‘house’ prices as opposed to ‘dwelling’ prices (around one quarter of all homes are not detached houses—they are apartments, semis, and terraces);
* comparing capital city house prices to incomes deriving from households located in all metro and non-metro regions (incomes are higher in the capitals);
* comparing dated 2007 median income estimates from the ABS to house prices in 2009 (the ABS provides much more timely quarterly income data via the National Accounts); and
* using wages rather than total disposable household incomes, where the latter includes more than one worker per household, and earnings on savings and investments (people buy homes out of disposable incomes rather than wages).
Rismark’s home price-to-income ratio index replicates the methodology recently used by the Governor of the RBA with two modifications: first, Rismark makes the comparison of home prices located in all regions across Australia with the incomes of households situated in those regions (the RBA contrasted capital city home prices with all-regions incomes); and, secondly, Rismark is now able to compare ‘average’ (as opposed to ‘median’) home prices with ‘average’ incomes.
The home sales data used in Rismark’s index derive from Australia’s most comprehensive residential property database (sourced exclusively from RP Data), which captures 100 per cent of all sales transacted across the country.
It is notable how Australia’s home price growth has generally tracked disposable incomes. In contrast to claims that Australian home prices are 7-8 times incomes, Rismark’s index suggests that the true ratio across all regions and all dwelling types is nearly half this estimate. This in turn implies that Australian housing is not as expensive as is commonly presumed.
Given that there is around $1 trillion of mortgage debt outstanding in Australia, which is secured against approximately $3.5 trillion of residential property, the debt-to-assets ratio (or gearing) is actually less than 30 per cent.
More importantly, recent RBA analysis has shown that despite Australia’s internationally high mortgage rates, debt-servicing is strong: Australia’s mortgage default rate is nearly one-tenth and one-quarter of US and UK levels, respectively.
Based on the RBA’s analysis at the end of last year, less than 40,000 of the circa 5 million borrowers were behind by 90 days or more on their repayments.
The information above has been extracted from the proprietary RP Data-Rismark Monthly publication and reproduced with permission.
*The average calculated by Rismark is a ‘trimmed mean’ similar to the measure used by the RBA when estimating inflation. A simple average results in volatile estimates due to occasional valuer general data entry errors. The trimming process results in the truncation of only the top and bottom 5 per cent of prices within the price distribution.
